From: "J. Bruce Fields" <bfields@fieldses.org>
To: bjschuma@netapp.com
Cc: linux-nfs@vger.kernel.org
Subject: Re: [PATCH 1/3] NFSD: Added fault injection
Date: Mon, 17 Oct 2011 18:18:11 -0400 [thread overview]
Message-ID: <20111017221810.GG13368@fieldses.org> (raw)
In-Reply-To: <1318009468-19893-1-git-send-email-bjschuma@netapp.com>
On Fri, Oct 07, 2011 at 01:44:26PM -0400, bjschuma@netapp.com wrote:
> diff --git a/fs/nfsd/Kconfig b/fs/nfsd/Kconfig
> index 10e6366..52fdd1c 100644
> --- a/fs/nfsd/Kconfig
> +++ b/fs/nfsd/Kconfig
> @@ -80,3 +80,13 @@ config NFSD_V4
> available from http://linux-nfs.org/.
>
> If unsure, say N.
> +
> +config NFSD_FAULT_INJECTION
> + bool "NFS server manual fault injection"
> + depends on NFSD_V4 && DEBUG_KERNEL
> + help
> + This option enables support for manually injectiong faults
^^^^^^^^^^
> diff --git a/fs/nfsd/fault_inject.c b/fs/nfsd/fault_inject.c
> new file mode 100644
> index 0000000..9f6815b
> --- /dev/null
> +++ b/fs/nfsd/fault_inject.c
> @@ -0,0 +1,94 @@
> +/*
> + * Copyright (c) 2011 Bryan Schumaker <bjschuma@netapp.com>
> + *
> + * Uses debugfs to create fault injection points for client testing
> + */
> +
> +#include <linux/types.h>
> +#include <linux/fs.h>
> +#include <linux/debugfs.h>
> +#include <linux/module.h>
> +
> +#include "state.h"
> +#include "fault_inject.h"
> +
> +struct nfsd_fault_inject_op {
> + char *action;
> + char *item;
> + char *file;
> + int file_data;
> + void (*func)(u64);
> +};
> +
> +#define INJECTION_OP(op_action, op_item, op_func) \
> +{ \
> + .action = op_action, \
> + .item = op_item, \
> + .file = op_action"_"op_item, \
> + .func = op_func, \
> +}
> +
> +static struct nfsd_fault_inject_op inject_ops[] = {
> + INJECTION_OP("forget", "clients", nfsd_forget_clients),
> + INJECTION_OP("forget", "locks", nfsd_forget_locks),
> + INJECTION_OP("forget", "openowners", nfsd_forget_openowners),
> + INJECTION_OP("forget", "delegations", nfsd_forget_delegations),
> + INJECTION_OP("recall", "delegations", nfsd_recall_delegations),
This is a little clever for my taste.... Could we just do
static struct nfsd_fault_inject_op inject_ops[] = {
{
.file = "forget_client",
.op = nfsd_forget_clients,
},
...
}
and do away with the separate item and action fields?
I'd rather be sort of obvious and boring even if it's slightly less
compact.
> +};
> +
> +static long int NUM_INJECT_OPS = sizeof(inject_ops) / sizeof(struct nfsd_fault_inject_op);
> +static struct dentry *debug_dir;
> +
> +static int nfsd_inject_set(void *data, u64 val)
> +{
> + int i;
> + struct nfsd_fault_inject_op *op;
> +
> + for (i = 0; i < NUM_INJECT_OPS; i++) {
> + op = &inject_ops[i];
> + if (&op->file_data == data) {
Huh, OK, so if I understand right, the contents of file_data doesn't
matter, you're just using a pointer to that field as a way to identify
the op array.
But then couldn't you just pass in a pointer to the op itself:
> + for (i = 0; i < NUM_INJECT_OPS; i++) {
> + op = &inject_ops[i];
> + debugfs_create_file(op->file, mode, debug_dir, &op->file_data, &fops_nfsd);
like:
debugfs_create_file(op->file, mode, debug_dir, op, &fops_nfsd);
and eliminate the file_data field?
Patches look OK otherwise on a quick skim, thanks.
--b.
> + }
> + return 0;
next prev parent reply other threads:[~2011-10-17 22:18 UTC|newest]
Thread overview: 14+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-10-07 17:44 [PATCH 1/3] NFSD: Added fault injection bjschuma
2011-10-07 17:44 ` [PATCH 2/3] NFSD: Added fault injection script bjschuma
2011-10-07 17:44 ` [PATCH 3/3] NFSD: Added fault injection documentation bjschuma
2011-10-17 22:18 ` J. Bruce Fields [this message]
2011-10-17 22:57 ` [PATCH 1/3] NFSD: Added fault injection Bryan Schumaker
2011-10-18 17:43 ` J. Bruce Fields
-- strict thread matches above, loose matches on Subject: below --
2011-10-24 11:20 bjschuma
2011-10-28 21:24 ` J. Bruce Fields
2011-11-01 14:08 ` Bryan Schumaker
2011-11-01 14:17 ` J. Bruce Fields
2011-11-01 14:22 ` Bryan Schumaker
2011-11-01 14:35 ` J. Bruce Fields
2011-11-01 17:35 bjschuma
2011-11-04 21:30 ` J. Bruce Fields
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20111017221810.GG13368@fieldses.org \
--to=bfields@fieldses.org \
--cc=bjschuma@netapp.com \
--cc=linux-nfs@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).